HORIBA RDE+ Application: Whole Vehicle RDE Compliance Solution. Steven (MIRA).pdf · Vehicles and...
Transcript of HORIBA RDE+ Application: Whole Vehicle RDE Compliance Solution. Steven (MIRA).pdf · Vehicles and...
HORIBA RDE+ Application:
Whole Vehicle RDE Compliance Solution Pune, India - 15th November 2019
1
Cleaner IC Engines for Sustainable Environment with
Innovative Emission Control Technologies (ECT 2019)
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in the-Loop
- Virtual Tools
• Conclusions
2
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
3
© HORIBA MIRA Ltd. 2019
Introduction
Regulatory Standards
• To regulate light-duty vehicle emissions, the WLTP and WLTC was implemented throughout Europe in
late 2017, replacing the NEDC.
• This was joined by the supplementary on-road procedure called RDE utilising PEMS.
• RDE regulations cover altitude, temperature and driving style as well as numerous other parameters.
4
© HORIBA MIRA Ltd. 2019
Introduction
What can be done to reduce Time, Cost and Risk?
Concept Development Certification SOP
Effo
rt a
nd
Co
st
Time
YesterdayWLTP Development
Concept Development with RDE Certification with RDE SOP
Increased Effort Increased Time
Effo
rt a
nd
Co
st
Time
TodayWLTP + RDE Development
Risk
& Cost
Concept Development Certification SOP
Increased Effort
Reduced Effort
Reduced TimeEffo
rt a
nd
Co
st
Time
TomorrowRDE+ Development
Risk
& Cost
5
© HORIBA MIRA Ltd. 2019
Introduction
RDE+ Application Portfolio
6
RDE + Product Portfolio
Architecture
STARS Enterprise
Software
Service
HardwareRDE Road Test
RDE Chassis TestRDE Engine/PT
TestRDE Virtual
Tools & HiL
Level 1Road
Level 2Chassis
Level 3Engine
Level 4VTs
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
7
© HORIBA MIRA Ltd. 2019
Methodology – Road Testing
Vehicles and Instrumentation
Euro Vehicle
Segment
B
(2016)
C
(2015)
M
(2017)
Body Type 5 Door Hatchback 5 Door MPV
Engine (Power)
Gasoline, 3 Cyl,
998cc,
Turbocharged
(74kW)
Diesel, 4 Cyl, 1499cc,
Turbocharged (88kW)
Diesel, 4 Cyl,
1461cc,
Turbocharged,
48v Hybrid
(81kW)
Transmission Manual 5 Spd Manual 6 Spd
EU Emission
StandardEURO 6b
Aftertreatment TWC EGR + DOC/DPF/LNTEGR +
DOC/DPF/LNT
Mass in Service
[kg]1130 1399 1583
PEMSHORIBA OBS ONE GS12 Gas Analyser & Particle Number
Unit
Amb. Temp.
Rel. HumidityHORIBA OBS ONE Weather Station
Altitude [m] HORIBA OBS ONE GPS
Base
Instrumentation
National Instruments CompactDAQ System: NI
9185/9862/9214/9205
Driveshaft
Strain GaugesAstech Electonics Rotary Telemetry System (RE3D)
8
© HORIBA MIRA Ltd. 2019
Methodology – Road Testing
Routes and ParametersR
DE
Te
sti
ng
Ph
ase
Lo
cati
on
Ta
rget
Te
mp
era
ture
[°C
]
Av
era
ge A
ltit
ud
e
[m]
To
tal/U
rba
n
Cu
mu
lati
ve
Po
sit
ive E
leva
tio
n
(CP
E)
[m/1
00k
m]
Dis
tan
ce S
plit
[%]
(Urb
an
/Ru
ral/M
Wa
y)
1Innsbruck,
Austria-7-0 623.2 498.1/579 32.5/31.2/36.3
2
3
Nuneaton,
UK
0-10
10-20105.1 491.0/611.6 37.0/33.8/29.2
4Vera,
Spain30-35 103.9 837.1/950.4 39.0/30.4/30.6
5Avila,
Spain30-35 1137.9 953.1/1022.4 33.3/30.3/36.4
• Driver “aggressivity” characterised by
RDE regulation parameter va_pos[95]
as a percentage of its limit:
- Gentle < 50%
- Moderate ≥ 50% ≤ 70%
- Aggressive > 70%
9
© HORIBA MIRA Ltd. 2019
Methodology – Road Testing
Emissions Results Overview
• All RDE emissions data is normalised to
corresponding WLTC emissions data (diamond
icon); expressed as the ratio of RDE to WLTC
emissions minus one.
• Significant variations in NOx and PN over full
range of RDE conditions.
• CO2 trend as expected – more aggressive driving,
more work done.
• Typical results for the gasoline vehicle are shown.
• More details of the road testing and associated
results is reported in SAE Technical Paper 2019-
01-0756.
10NOx is typically higher for gentle drives compared to
aggressive drives on the same route (opposite trend for PN)Possibly caused by “light-out” of
the TWC during gentle drives?
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
11
© HORIBA MIRA Ltd. 2019
Methodology – Chassis Dynamometer Testing
Chassis Dynamometer Specification and Equipment
Equipment
Analyser Room
HORIBA Supplied:
DLT and DLS dilution tunnel and sampler
CVS-ONE constant volume sampler
MEXA ONE D2 EGR raw exhaust gas sampler
MEXA ONE C1 SL OV dilute exhaust gas (bag) analyser
MEXA 2000 SPCS solid particle counter
MEXA ONE QL NX Quantum cascade laser system
GMC ONE particulate measurement
PFS ONE robotised particulate filter weighing system
OBS ONE PEMS kit (gaseous and particle)
Test Cell
FWD, RWD, AWD
-20°C to 35°C (plus 3 climatic soak chambers)
30-60% relative humidity
Vehicle cooling fan
Dynamometer
Data
230kW per axle
300km/h maximum speed
12
© HORIBA MIRA Ltd. 2019
Methodology – Chassis Dynamometer Testing
Chassis Dynamometer Specification and Equipment
HORIBA Multi-Function Efficient Dynamic Altitude Simulation (MEDAS) unit will be used for altitude and ambient temperature requirements for RDE+
13
© HORIBA MIRA Ltd. 2019
Methodology – Vehicle Load Replication
Highly instrumented vehicle to measure loads
• To accelerate the R2R RDE+ methodology, the chassis
dynamometer testing focused on the gasoline vehicle:
simple aftertreatment, sea-level driving route (Nuneaton,
UK).
• Vehicle load measured directly via strain gauges on
drive shafts
• “Effective road gradient” derived by subtracting
coastdown terms and inertia from the measured road
data – programmed into chassis dynamometer with
vehicle speed and gear selection.
• Human driver used throughout initial RDE chassis
dynamometer drive.
Work
[MJ]
CO
[g]
CO2
[g]
NOx
[g]
PN
[#]
Road 52.1 43.1 14830 22.97 2.13E+14
Chassis 54.1 44.3 15139 21.30 1.90E+14
Delta [%] 3.8 2.8 2.1 -7.3 -10.8
14
© HORIBA MIRA Ltd. 2019
Methodology – Vehicle Load Replication
Human Driver vs. Robot Driver (1)
• To improve repeatability and durability of the ~90 minute RDE cycle replication, a robot driver has been
used for further testing. The robot has been trained to follow the cycle, change gear and brake.
• The robot was more accurate, repeatable and durable.
• Overall, a very good correlation was made with the human driver.
15
© HORIBA MIRA Ltd. 2019
Methodology – Vehicle Load Replication
High Altitude Route, Robot Driver, Initial Results (1)
• Replication of high altitude (Innsbruck, Austria)
route using HORIBA MEDAS system & robot
driver.
• Excellent correlation achieved with measured
RDE route data.
• Similar level of correlation with sea-level route.
Work
[MJ]
CO
[g]
CO2
[g]
NOx
[g]
PN
[#]
Road 43.8 14.4 12408 23.0 1.54E+14
Chassis 44.4 18.8 12503 24.8 1.57E+14
Delta [%] 1.4 30.6* 0.8 7.8 1.9
16
* Issue with PEMS CO analyser
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
17
© HORIBA MIRA Ltd. 2019
© 2019 HORIBA, Ltd. All rights reserved
Traditional Track to Lab Model (Assumption):
𝐹 𝑡 = 𝐴 + 𝐵 ∗ 𝑣 𝑡 + 𝐶 ∗ 𝑣 𝑡 2 +𝑀 ∗𝑑𝑣 𝑡
𝑑𝑡But the real world has variable road grade:
𝐹 𝑡 = 𝐴 + 𝐵 ∗ 𝑣 𝑡 + 𝐶 ∗ 𝑣 𝑡 2 +𝑀 ∗𝑑𝑣 𝑡
𝑑𝑡+ 𝑀 ∗ 𝑔 ∗ 𝑠𝑖𝑛⍺ 𝑡
But the real world has variable winds:
𝐹 𝑡 = 𝐴 + 𝐵 ∗ 𝑣𝑔 𝑡 + 𝐶 ∗ 𝑣𝑎 𝑡 2 +𝑀 ∗𝑑𝑣𝑔 𝑡
𝑑𝑡+𝑀 ∗ 𝑔 ∗ 𝑠𝑖𝑛⍺ 𝑡
But the real world has variable air densities:
𝐹 𝑡 = 𝐴 + 𝐵 ∗ 𝑣𝑔 𝑡 + 𝜌 Τ𝑡 𝜌0 ∗ 𝐶 ∗ 𝑣𝑎 𝑡 2 +𝑀 ∗𝑑𝑣𝑔 𝑡
𝑑𝑡+ 𝑀 ∗ 𝑔 ∗ 𝑠𝑖𝑛⍺ 𝑡
But this road load equation is still an idealization, road surfaces change from
block to block, cornering affects load, and road grade is difficult to measure precisely
with high resolution:
There must be a better way…
Replicating Real World Load Can Be Complicated
18
© HORIBA MIRA Ltd. 2019
… a better way, HORIBA “Torque Matching”
© 2019 HORIBA, Ltd. All rights reserved
Forget all the load modeling and abstractions
Road Test
• Record speed, pedal, weather
conditions
• Any grade, surface, weather,
altitude, and cornering
Lab Validation Test
(Road Test Replication)
• MEDAS matches ambient
• Dyno matches speed
• Robot driver matches pedal
• Dyno torque recorded
Lab Tests
(Road Test Simulations)
• MEDAS matches ambient
• Dyno matches speed
• Robot driver matches torque
• Change vehicle calibration or
emission controls as desired and
repeat test
2. 3+…1.
19
© HORIBA MIRA Ltd. 2019
Torque Matching Simulates Real World Load
✔✔
✔✔
© 2019 HORIBA, Ltd. All rights reserved
✔
✔
✔✔
✔
✔
• No torque wheels or instrumented drive shafts
• No coastdowns
• No road grade measurement or estimation
• Blind testing is enabled
• Lab testing precision is enabled
• Impact of changes to vehicle or calibration can be
assessed with back to back, repeatable testing
Torque Matching on a Chassis Dynamometer
Powertrain Calibration Space
✔
20
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
21
© HORIBA MIRA Ltd. 2019
Methodology – Engine-in-the-Loop
Equipment
Equipment
Analysers
MEXA ONE D2 EGR exhaust gas analysis system
OBS ONE PEMS GS12 kit (gaseous and particle)
MEXA-2100SPCS Real Time Particle Counter
MEXA ONE QL NX Quantum cascade laser system
Test Cell
HORIBA DYNASPM LI 470 AC Dyno
Hot and cold box (engine containment) -30°C to 50°C
HORIBA MEDAS – temperature, pressure and humidity
Misc
AVL Indicom X-Ion high-speed data acquisition
ETAS INCA
HORIBA STARS SPARC
HORIBA STARS Calibrate (pseudo-automapping
functionality)
IPG CarMaker virtual simulation tool (vehicle and
driving style)
22
© HORIBA MIRA Ltd. 2019
Methodology – Engine-in-the-Loop
EiL Setup (1)
VEHICLE, ROAD, DRIVER &
ENVIRONMENTAL DATA
CREATION OF
VIRTUAL
ENVIRONMENT
ENGINE SPEED AND
LOAD/PEDAL CLOSED
LOOP CONTROL
COMPLETE ROAD, ENVIRONMENTAL AND VEHICLE REPLICATION
23
© HORIBA MIRA Ltd. 2019
Methodology – Engine-in-the-Loop
EiL Setup (2)
• IPG CarMaker to STARS interface successfully installed – purchased vehicle, no OEM support.
• Currently running through Nuneaton RDE cycle with different driving modes.
Torque Demand
Speed Demand
24
Closed-loop load control through IPG
CarMaker
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
25
© HORIBA MIRA Ltd. 2019
Gentle
Methodology – Engine-in-the-Loop
Sea Level Route, EiL Initial Results (1)
• Effects of driving style on vehicle and engine
performance across Nuneaton, UK RDE route
assessed using IPG CarMaker.
• 5 parameter DOE – optimising for va_pos[95]
% of limit – HORIBA MIRA definition of driving
style.
- Longitudinal and lateral acceleration
- Longitudinal deceleration
- Pedal dynamics
• 3 cycles run across fixed route
- Gentle drive – va_pos[95] % of limit = 40%
- Normal drive – va_pos[95] % of limit = 70%
- Aggressive drive – va_pos[95] % of limit = 90
26
Gentle NormalGentle Normal Aggressive
© HORIBA MIRA Ltd. 2019
Methodology – Engine-in-the-Loop
Sea Level Route, EiL Initial Results (2)
• Rapid screening of calibration and
hardware changes without needing to
schedule on-road RDE testing.
• Run repeat RDEs without influences such
as traffic, weather and other anomalies.
• Identify problematic areas of the engine
operating map that are often not related to
time residency.
• Resolve cycle emissions before prototype
testing.
27
© HORIBA MIRA Ltd. 2019
Methodology – Engine-in-the-Loop
Sea Level Route, EiL Initial Results (2)
28
Tailpipe CO2
[g/km]
Tailpipe NOx
[mg/km]
Tailpipe PN
[#/km]
Fuel Consump
[l/100km]
Fuel Consump
[mpg]
Gentle 136.7 32.1 4.8E+11 6.1 46.2
Normal 153.0 53.0 5.3E+11 6.8 41.4
Aggressive 170.1 100.3 1.1E+12 7.5 37.7
© HORIBA MIRA Ltd. 2019
Presentation Outline
• Introduction
• Methodology
- Road Testing
- Chassis Dynamometer
o Vehicle load replication (complex method)
o “Torque Matching” (new unique HORIBA simplified method)
- Engine-in-the-Loop
- Virtual Tools
• Conclusions
29
© HORIBA MIRA Ltd. 2019
Conclusions
• HORIBA MIRA’s R2R RDE+ test methodology has been described.
• The road part of the programme is complete with work underway on the chassis dynamometer and EiL
segments.
• Sea-level and high altitude, cold temperature RDE routes have been successfully correlated with the
vehicle driven on the chassis dynamometer utilising a robot driver and HORIBA MEDAS altitude
emulation device.
• The effects of driving style on engine performance and emissions for a fixed RDE route have been
presented using an EiL toolchain.
• The EiL methodology will allow OEMs to front-load powertrain design, development and calibration
activities thus resulting in fewer prototype vehicles and physical climatic testing to achieve RDE
compliance.
• By adopting road, chassis, EiL and virtual testing (RDE+), many of the unknown scenarios that arise
through real testing can be mitigated much further upstream; thus reducing time, effort, money and
pollution.
30
© HORIBA MIRA Ltd. 2019
Danke
Большое спасибо
Grazie
Σας ευχαριστούμε
감사합니다
Obrigado谢谢
ขอบคุณครับ
ありがとうございました
धन्यवादநன்ற
Cảm ơn
Dziękuję
Tack ska ni ha
Thank you
Merci
Gracias
31